Mitsuki-Sakamoto/alpaca_farm-deberta-re-pref-64-_fil_self_160m_bo16_2_mix_50_kl_0.1_prm_160m_thr_0.0_seed_2

--- dataset_info: config_name: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1 features: - name: instruction dtype: string - name: input dtype: string - name: output dtype: string - name: preference dtype: int64 - name: output_1 dtype: string - name: output_2 dtype: string - name: reward_model_prompt_format dtype: string - name: gen_prompt_format dtype: string - name: gen_kwargs struct: - name: do_sample dtype: bool - name: max_new_tokens dtype: int64 - name: pad_token_id dtype: int64 - name: top_k dtype: int64 - name: top_p dtype: float64 - name: reward_1 dtype: float64 - name: reward_2 dtype: float64 - name: n_samples dtype: int64 - name: reject_select dtype: string - name: index dtype: int64 - name: prompt dtype: string - name: chosen dtype: string - name: rejected dtype: string - name: filtered_epoch dtype: int64 - name: gen_reward dtype: float64 - name: gen_response dtype: string splits: - name: epoch_0 num_bytes: 43800231 num_examples: 18928 - name: epoch_1 num_bytes: 44360861 num_examples: 18928 - name: epoch_2 num_bytes: 44434638 num_examples: 18928 - name: epoch_3 num_bytes: 44453497 num_examples: 18928 - name: epoch_4 num_bytes: 44453617 num_examples: 18928 - name: epoch_5 num_bytes: 44449467 num_examples: 18928 - name: epoch_6 num_bytes: 44436339 num_examples: 18928 - name: epoch_7 num_bytes: 44425307 num_examples: 18928 - name: epoch_8 num_bytes: 44419489 num_examples: 18928 - name: epoch_9 num_bytes: 44416273 num_examples: 18928 - name: epoch_10 num_bytes: 44413967 num_examples: 18928 - name: epoch_11 num_bytes: 44411729 num_examples: 18928 - name: epoch_12 num_bytes: 44409934 num_examples: 18928 - name: epoch_13 num_bytes: 44409443 num_examples: 18928 - name: epoch_14 num_bytes: 44407291 num_examples: 18928 - name: epoch_15 num_bytes: 44406704 num_examples: 18928 - name: epoch_16 num_bytes: 44406563 num_examples: 18928 - name: epoch_17 num_bytes: 44405602 num_examples: 18928 - name: epoch_18 num_bytes: 44407357 num_examples: 18928 - name: epoch_19 num_bytes: 44405259 num_examples: 18928 - name: epoch_20 num_bytes: 44407013 num_examples: 18928 - name: epoch_21 num_bytes: 44407420 num_examples: 18928 - name: epoch_22 num_bytes: 44406446 num_examples: 18928 - name: epoch_23 num_bytes: 44406607 num_examples: 18928 - name: epoch_24 num_bytes: 44404891 num_examples: 18928 - name: epoch_25 num_bytes: 44405934 num_examples: 18928 - name: epoch_26 num_bytes: 44407169 num_examples: 18928 - name: epoch_27 num_bytes: 44404320 num_examples: 18928 - name: epoch_28 num_bytes: 44406013 num_examples: 18928 - name: epoch_29 num_bytes: 44404414 num_examples: 18928 download_size: 701318354 dataset_size: 1331793795 configs: - config_name: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1 data_files: - split: epoch_0 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_0-* - split: epoch_1 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_1-* - split: epoch_2 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_2-* - split: epoch_3 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_3-* - split: epoch_4 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_4-* - split: epoch_5 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_5-* - split: epoch_6 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_6-* - split: epoch_7 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_7-* - split: epoch_8 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_8-* - split: epoch_9 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_9-* - split: epoch_10 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_10-* - split: epoch_11 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_11-* - split: epoch_12 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_12-* - split: epoch_13 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_13-* - split: epoch_14 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_14-* - split: epoch_15 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_15-* - split: epoch_16 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_16-* - split: epoch_17 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_17-* - split: epoch_18 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_18-* - split: epoch_19 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_19-* - split: epoch_20 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_20-* - split: epoch_21 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_21-* - split: epoch_22 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_22-* - split: epoch_23 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_23-* - split: epoch_24 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_24-* - split: epoch_25 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_25-* - split: epoch_26 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_26-* - split: epoch_27 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_27-* - split: epoch_28 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_28-* - split: epoch_29 path: alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_29-* ---数据集信息： 配置名称：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1 数据字段： - 字段名：instruction，数据类型：字符串，代表用户指令 - 字段名：input，数据类型：字符串，代表输入上下文 - 字段名：output，数据类型：字符串，代表基础输出结果 - 字段名：preference，数据类型：64位整数，代表偏好标签 - 字段名：output_1，数据类型：字符串，代表候选输出1 - 字段名：output_2，数据类型：字符串，代表候选输出2 - 字段名：reward_model_prompt_format，数据类型：字符串，代表奖励模型提示模板格式 - 字段名：gen_prompt_format，数据类型：字符串，代表生成任务提示模板格式 - 字段名：gen_kwargs，数据类型：结构体，为生成超参数配置，包含子字段： - 字段名：do_sample，数据类型：布尔值，控制是否开启采样 - 字段名：max_new_tokens，数据类型：64位整数，指定最大生成Token（Token）数 - 字段名：pad_token_id，数据类型：64位整数，指定填充Token（Token）的ID - 字段名：top_k，数据类型：64位整数，指定Top-K采样参数 - 字段名：top_p，数据类型：双精度浮点数，指定Top-P采样参数 - 字段名：reward_1，数据类型：双精度浮点数，代表候选输出1的奖励得分 - 字段名：reward_2，数据类型：双精度浮点数，代表候选输出2的奖励得分 - 字段名：n_samples，数据类型：64位整数，代表样本数量 - 字段名：reject_select，数据类型：字符串，代表拒绝选择标记 - 字段名：index，数据类型：64位整数，代表数据索引 - 字段名：prompt，数据类型：字符串，代表完整提示文本 - 字段名：chosen，数据类型：字符串，代表被选中的优质回复 - 字段名：rejected，数据类型：字符串，代表被拒绝的劣质回复 - 字段名：filtered_epoch，数据类型：64位整数，代表过滤轮次 - 字段名：gen_reward，数据类型：双精度浮点数，代表生成回复的奖励得分 - 字段名：gen_response，数据类型：字符串，代表模型生成的回复 划分： - 划分名称：epoch_0，字节大小：43800231，样本数量：18928 - 划分名称：epoch_1，字节大小：44360861，样本数量：18928 - 划分名称：epoch_2，字节大小：44434638，样本数量：18928 - 划分名称：epoch_3，字节大小：44453497，样本数量：18928 - 划分名称：epoch_4，字节大小：44453617，样本数量：18928 - 划分名称：epoch_5，字节大小：44449467，样本数量：18928 - 划分名称：epoch_6，字节大小：44436339，样本数量：18928 - 划分名称：epoch_7，字节大小：44425307，样本数量：18928 - 划分名称：epoch_8，字节大小：44419489，样本数量：18928 - 划分名称：epoch_9，字节大小：44416273，样本数量：18928 - 划分名称：epoch_10，字节大小：44413967，样本数量：18928 - 划分名称：epoch_11，字节大小：44411729，样本数量：18928 - 划分名称：epoch_12，字节大小：44409934，样本数量：18928 - 划分名称：epoch_13，字节大小：44409443，样本数量：18928 - 划分名称：epoch_14，字节大小：44407291，样本数量：18928 - 划分名称：epoch_15，字节大小：44406704，样本数量：18928 - 划分名称：epoch_16，字节大小：44406563，样本数量：18928 - 划分名称：epoch_17，字节大小：44405602，样本数量：18928 - 划分名称：epoch_18，字节大小：44407357，样本数量：18928 - 划分名称：epoch_19，字节大小：44405259，样本数量：18928 - 划分名称：epoch_20，字节大小：44407013，样本数量：18928 - 划分名称：epoch_21，字节大小：44407420，样本数量：18928 - 划分名称：epoch_22，字节大小：44406446，样本数量：18928 - 划分名称：epoch_23，字节大小：44406607，样本数量：18928 - 划分名称：epoch_24，字节大小：44404891，样本数量：18928 - 划分名称：epoch_25，字节大小：44405934，样本数量：18928 - 划分名称：epoch_26，字节大小：44407169，样本数量：18928 - 划分名称：epoch_27，字节大小：44404320，样本数量：18928 - 划分名称：epoch_28，字节大小：44406013，样本数量：18928 - 划分名称：epoch_29，字节大小：44404414，样本数量：18928 下载总大小：701318354 字节 数据集总占用大小：1331793795 字节 配置： - 配置名称：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1 数据文件： - 划分：epoch_0，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_0-* - 划分：epoch_1，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_1-* - 划分：epoch_2，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_2-* - 划分：epoch_3，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_3-* - 划分：epoch_4，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_4-* - 划分：epoch_5，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_5-* - 划分：epoch_6，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_6-* - 划分：epoch_7，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_7-* - 划分：epoch_8，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_8-* - 划分：epoch_9，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_9-* - 划分：epoch_10，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_10-* - 划分：epoch_11，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_11-* - 划分：epoch_12，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_12-* - 划分：epoch_13，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_13-* - 划分：epoch_14，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_14-* - 划分：epoch_15，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_15-* - 划分：epoch_16，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_16-* - 划分：epoch_17，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_17-* - 划分：epoch_18，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_18-* - 划分：epoch_19，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_19-* - 划分：epoch_20，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_20-* - 划分：epoch_21，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_21-* - 划分：epoch_22，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_22-* - 划分：epoch_23，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_23-* - 划分：epoch_24，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_24-* - 划分：epoch_25，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_25-* - 划分：epoch_26，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_26-* - 划分：epoch_27，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_27-* - 划分：epoch_28，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_28-* - 划分：epoch_29，路径：alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_29-*

原始信息汇总

数据集概述

数据集配置名称

• alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1

数据集特征

• instruction: 数据类型为字符串
• input: 数据类型为字符串
• output: 数据类型为字符串
• preference: 数据类型为int64
• output_1: 数据类型为字符串
• output_2: 数据类型为字符串
• reward_model_prompt_format: 数据类型为字符串
• gen_prompt_format: 数据类型为字符串
• gen_kwargs: 结构化数据，包含以下字段：
  • do_sample: 数据类型为布尔值
  • max_new_tokens: 数据类型为int64
  • pad_token_id: 数据类型为int64
  • top_k: 数据类型为int64
  • top_p: 数据类型为float64
• reward_1: 数据类型为float64
• reward_2: 数据类型为float64
• n_samples: 数据类型为int64
• reject_select: 数据类型为字符串
• index: 数据类型为int64
• prompt: 数据类型为字符串
• chosen: 数据类型为字符串
• rejected: 数据类型为字符串
• filtered_epoch: 数据类型为int64
• gen_reward: 数据类型为float64
• gen_response: 数据类型为字符串

数据集分割

• epoch_0 至 epoch_29:
  • 每个分割的示例数量均为18928
  • 每个分割的字节数在43800231至44453617之间

数据集大小

• 下载大小: 701318354字节
• 数据集总大小: 1331793795字节

数据文件配置

• 每个分割对应的路径格式为alpaca_instructions-pythia_160m_alpaca_farm_instructions_sft_constant_pa_seed_1/epoch_{n}-*，其中n为0至29。